def test_generate_sample_less_arg(self, random_state, sample_discrete):
model = MC(['a', 'b'], [2, 2])
model.transition_models['a'] = {
0: {
0: 0.1,
1: 0.9
},
1: {
0: 0.2,
1: 0.8
}
}
model.transition_models['b'] = {
0: {
0: 0.3,
1: 0.7
},
1: {
0: 0.4,
1: 0.6
}
}
random_state.return_value = [State('a', 0), State('b', 1)]
sample_discrete.side_effect = [[1], [0]] * 2
gen = model.generate_sample(size=2)
samples = [sample for sample in gen]
expected_samples = [[State('a', 1), State('b', 0)]] * 2
self.assertEqual(samples, expected_samples)
def test_set_start_state_list(self, check_state):
model = MC(['b', 'a'], [1, 2])
check_state.return_value = True
model.set_start_state([State('a', 0), State('b', 1)])
model_state = [State('b', 1), State('a', 0)]
check_state.assert_called_once_with(model, model_state)
self.assertEqual(model.state, model_state)
def test_sample(self):
model = MC(['a', 'b'], [2, 2])
model.transition_models['a'] = {
0: {
0: 0.1,
1: 0.9
},
1: {
0: 0.2,
1: 0.8
}
}
model.transition_models['b'] = {
0: {
0: 0.3,
1: 0.7
},
1: {
0: 0.4,
1: 0.6
}
}
sample = model.sample(start_state=[State('a', 0),
State('b', 1)],
size=2)
self.assertEqual(len(sample), 2)
self.assertEqual(list(sample.columns), ['a', 'b'])
self.assertTrue(
list(sample.loc[0]) in [[0, 0], [0, 1], [1, 0], [1, 1]])
self.assertTrue(
list(sample.loc[1]) in [[0, 0], [0, 1], [1, 0], [1, 1]])
def test_sample_less_arg(self, random_state):
model = MC(['a', 'b'], [2, 2])
random_state.return_value = [State('a', 0), State('b', 1)]
sample = model.sample(size=1)
random_state.assert_called_once_with(model)
self.assertEqual(model.state, random_state.return_value)
self.assertEqual(len(sample), 1)
self.assertEqual(list(sample.columns), ['a', 'b'])
self.assertEqual(list(sample.loc[0]), [0, 1])
def test_sample_less_arg(self, random_state):
self.gibbs.state = None
random_state.return_value = [
State('diff', 0),
State('intel', 0),
State('grade', 0)
]
sample = self.gibbs.sample(size=2)
random_state.assert_called_once_with(self.gibbs)
self.assertEqual(len(sample), 2)
def test_generate_sample(self):
start_state = [State('diff', 0), State('intel', 0), State('grade', 0)]
gen = self.gibbs.generate_sample(start_state, 2)
samples = [sample for sample in gen]
self.assertEqual(len(samples), 2)
self.assertEqual(
{samples[0][0].var, samples[0][1].var, samples[0][2].var},
{'diff', 'intel', 'grade'})
self.assertEqual(
{samples[1][0].var, samples[1][1].var, samples[1][2].var},
{'diff', 'intel', 'grade'})
def test_sample(self):
start_state = [State('diff', 0), State('intel', 0), State('grade', 0)]
sample = self.gibbs.sample(start_state, 2)
self.assertEquals(len(sample), 2)
self.assertEquals(len(sample.columns), 3)
self.assertIn('diff', sample.columns)
self.assertIn('intel', sample.columns)
self.assertIn('grade', sample.columns)
self.assertTrue(set(sample['diff']).issubset({0, 1}))
self.assertTrue(set(sample['intel']).issubset({0, 1}))
self.assertTrue(set(sample['grade']).issubset({0, 1, 2}))
def setUp(self):
self.variables = ['intel', 'diff', 'grade']
self.card = [3, 2, 3]
self.cardinalities = {'intel': 3, 'diff': 2, 'grade': 3}
self.intel_tm = {0: {0: 0.1, 1: 0.25, 2: 0.65}, 1: {0: 0.5, 1: 0.3, 2: 0.2}, 2: {0: 0.3, 1: 0.3, 2: 0.4}}
self.diff_tm = {0: {0: 0.3, 1: 0.7}, 1: {0: 0.75, 1: 0.25}}
self.grade_tm = {0: {0: 0.4, 1: 0.2, 2: 0.4}, 1: {0: 0.9, 1: 0.05, 2: 0.05}, 2: {0: 0.1, 1: 0.4, 2: 0.5}}
self.start_state = [State('intel', 0), State('diff', 1), State('grade', 2)]
self.model = MC()
self.sample = DataFrame(index=range(200), columns=['a', 'b'])
self.sample.a = [1] * 100 + [0] * 100
self.sample.b = [0] * 100 + [1] * 100
def set_start_state(self, start_state):
"""
Set the start state of the Markov Chain. If the start_state is given as a array-like iterable, its contents
are reordered in the internal representation.
Parameters:
-----------
start_state: dict or array-like iterable object
Dict (or list) of tuples representing the starting states of the variables.
Examples:
---------
>>> from pgm.models import MarkovChain as MC
>>> from pgm.factors.discrete import State
>>> model = MC(['a', 'b'], [2, 2])
>>> model.set_start_state([State('a', 0), State('b', 1)])
"""
if start_state is not None:
if not hasattr(start_state, '__iter__') or isinstance(
start_state, six.string_types):
raise ValueError('start_state must be a non-string iterable.')
# Must be an array-like iterable. Reorder according to self.variables.
state_dict = {var: st for var, st in start_state}
start_state = [
State(var, state_dict[var]) for var in self.variables
]
if start_state is None or self._check_state(start_state):
self.state = start_state
def test_copy(self):
model = MC(['a', 'b'], [2, 2], [State('a', 0), State('b', 1)])
model.add_transition_model('a', {0: {0: 0.1, 1: 0.9}, 1: {0: 0.2, 1: 0.8}})
model.add_transition_model('b', {0: {0: 0.3, 1: 0.7}, 1: {0: 0.4, 1: 0.6}})
copy = model.copy()
self.assertIsInstance(copy, MC)
self.assertEqual(sorted(model.variables), sorted(copy.variables))
self.assertEqual(model.cardinalities, copy.cardinalities)
self.assertEqual(model.transition_models, copy.transition_models)
self.assertEqual(model.state, copy.state)
model.add_variable('p', 1)
model.set_start_state([State('a', 0), State('b', 1), State('p', 0)])
model.add_transition_model('p', {0: {0: 1}})
self.assertNotEqual(sorted(model.variables), sorted(copy.variables))
self.assertEqual(sorted(['a', 'b']), sorted(copy.variables))
self.assertNotEqual(model.cardinalities, copy.cardinalities)
self.assertEqual({'a': 2, 'b': 2}, copy.cardinalities)
self.assertNotEqual(model.state, copy.state)
self.assertEqual([State('a', 0), State('b', 1)], copy.state)
self.assertNotEqual(model.transition_models, copy.transition_models)
self.assertEqual(len(copy.transition_models), 2)
self.assertEqual(copy.transition_models['a'], {0: {0: 0.1, 1: 0.9}, 1: {0: 0.2, 1: 0.8}})
self.assertEqual(copy.transition_models['b'], {0: {0: 0.3, 1: 0.7}, 1: {0: 0.4, 1: 0.6}})
def test_rejection_sample_basic(self):
sample = self.sampling_inference.rejection_sample(
[State('A', 1), State('J', 1),
State('R', 1)], 25)
self.assertEquals(len(sample), 25)
self.assertEquals(len(sample.columns), 6)
self.assertIn('A', sample.columns)
self.assertIn('J', sample.columns)
self.assertIn('R', sample.columns)
self.assertIn('Q', sample.columns)
self.assertIn('G', sample.columns)
self.assertIn('L', sample.columns)
self.assertTrue(set(sample.A).issubset({1}))
self.assertTrue(set(sample.J).issubset({1}))
self.assertTrue(set(sample.R).issubset({1}))
self.assertTrue(set(sample.Q).issubset({0, 1}))
self.assertTrue(set(sample.G).issubset({0, 1}))
self.assertTrue(set(sample.L).issubset({0, 1}))
def test_likelihood_weighted_sample(self):
sample = self.sampling_inference.likelihood_weighted_sample(
[State('A', 0), State('J', 1),
State('R', 0)], 25)
self.assertEquals(len(sample), 25)
self.assertEquals(len(sample.columns), 7)
self.assertIn('A', sample.columns)
self.assertIn('J', sample.columns)
self.assertIn('R', sample.columns)
self.assertIn('Q', sample.columns)
self.assertIn('G', sample.columns)
self.assertIn('L', sample.columns)
self.assertIn('_weight', sample.columns)
self.assertTrue(set(sample.A).issubset({0, 1}))
self.assertTrue(set(sample.J).issubset({0, 1}))
self.assertTrue(set(sample.R).issubset({0, 1}))
self.assertTrue(set(sample.Q).issubset({0, 1}))
self.assertTrue(set(sample.G).issubset({0, 1}))
self.assertTrue(set(sample.L).issubset({0, 1}))
def random_state(self):
"""
Generates a random state of the Markov Chain.
Return Type:
------------
List of namedtuples, representing a random assignment to all variables of the model.
Examples:
---------
>>> from pgm.models import MarkovChain as MC
>>> model = MC(['intel', 'diff'], [2, 3])
>>> model.random_state()
[State('diff', 2), State('intel', 1)]
"""
return [
State(var, np.random.randint(self.cardinalities[var]))
for var in self.variables
]
def generate_sample(self, start_state=None, size=1):
"""
Generator version of self.sample
Return Type:
------------
List of State namedtuples, representing the assignment to all variables of the model.
Examples:
---------
>>> from pgm.models.MarkovChain import MarkovChain
>>> from pgm.factors.discrete import State
>>> model = MarkovChain()
>>> model.add_variables_from(['intel', 'diff'], [3, 2])
>>> intel_tm = {0: {0: 0.2, 1: 0.4, 2:0.4}, 1: {0: 0, 1: 0.5, 2: 0.5}, 2: {0: 0.3, 1: 0.3, 2: 0.4}}
>>> model.add_transition_model('intel', intel_tm)
>>> diff_tm = {0: {0: 0.5, 1: 0.5}, 1: {0: 0.25, 1:0.75}}
>>> model.add_transition_model('diff', diff_tm)
>>> gen = model.generate_sample([State('intel', 0), State('diff', 0)], 2)
>>> [sample for sample in gen]
[[State(var='intel', state=2), State(var='diff', state=1)],
[State(var='intel', state=2), State(var='diff', state=0)]]
"""
if start_state is None:
if self.state is None:
self.state = self.random_state()
# else use previously-set state
else:
self.set_start_state(start_state)
# sampled.loc[0] = [self.state[var] for var in self.variables]
for i in range(size):
for j, (var, st) in enumerate(self.state):
next_st = sample_discrete(
list(self.transition_models[var][st].keys()),
list(self.transition_models[var][st].values()))[0]
self.state[j] = State(var, next_st)
yield self.state[:]
def test_prob_from_sample(self, sample):
model = MC(['a', 'b'], [2, 2])
sample.return_value = self.sample
probabilites = model.prob_from_sample([State('a', 1), State('b', 0)])
self.assertEqual(list(probabilites), [1] * 50 + [0] * 50)
def sample(self, start_state=None, size=1):
"""
Sample from the Markov Chain.
Parameters:
-----------
start_state: dict or array-like iterable
Representing the starting states of the variables. If None is passed, a random start_state is chosen.
size: int
Number of samples to be generated.
Return Type:
------------
pandas.DataFrame
Examples:
---------
>>> from pgm.models import MarkovChain as MC
>>> from pgm.factors.discrete import State
>>> model = MC(['intel', 'diff'], [2, 3])
>>> model.set_start_state([State('intel', 0), State('diff', 2)])
>>> intel_tm = {0: {0: 0.25, 1: 0.75}, 1: {0: 0.5, 1: 0.5}}
>>> model.add_transition_model('intel', intel_tm)
>>> diff_tm = {0: {0: 0.1, 1: 0.5, 2: 0.4}, 1: {0: 0.2, 1: 0.2, 2: 0.6 }, 2: {0: 0.7, 1: 0.15, 2: 0.15}}
>>> model.add_transition_model('diff', diff_tm)
>>> model.sample(size=5)
intel diff
0 0 2
1 1 0
2 0 1
3 1 0
4 0 2
"""
if start_state is None:
if self.state is None:
self.state = self.random_state()
# else use previously-set state
else:
self.set_start_state(start_state)
sampled = DataFrame(index=range(size), columns=self.variables)
sampled.loc[0] = [st for var, st in self.state]
var_states = defaultdict(dict)
var_values = defaultdict(dict)
samples = defaultdict(dict)
for var in self.transition_models.keys():
for st in self.transition_models[var]:
var_states[var][st] = list(
self.transition_models[var][st].keys())
var_values[var][st] = list(
self.transition_models[var][st].values())
samples[var][st] = sample_discrete(var_states[var][st],
var_values[var][st])[0]
for i in range(size - 1):
for j, (var, st) in enumerate(self.state):
next_st = samples[var][st]
self.state[j] = State(var, next_st)
sampled.loc[i + 1] = [st for var, st in self.state]
return sampled
def test_check_state_bad_vars(self):
model = MC()
# state_vars and model_vars differ
self.assertRaises(ValueError, model._check_state, [State(1, 2)])
def test_check_state_success(self):
model = MC(['a'], [2])
self.assertTrue(model._check_state([State('a', 1)]))
def test_check_state_bad_var_value(self):
model = MC(['a'], [2])
# value of variable >= cardinaliity
self.assertRaises(ValueError, model._check_state, [State('a', 3)])